Non-invasive diagnosis of human beta cell damage and death Abstract Detection of beta cell damage or death in blood is an important challenge that could transform basic research, diagnosis and clinical trials in diabetes. We propose to establish novel, sensitive and quantitative methods for the detection of nucleic acids shed from damaged beta cells to blood, and to apply these methods in the context of human diabetes. In one approach, we will define a unique DNA methylation signature of beta cells and utilize next generation sequencing technology to detect and quantify signature DNA fragments released from dying beta cells to the circulation. In preliminary studies we found that this approach can robustly detect beta cell death in the blood of recently diagnosed type 1 diabetes patients and in patients transplanted with islets, while healthy donors show an extremely low background. In a second approach, we will quantify microRNAs released from damaged or dying beta cells to the circulation, focusing on combinations of beta cell-specific microRNAs. Here as well, preliminary studies on recently diagnosed type 1 diabetes patients suggest that beta cell damage (lethal or sub-lethal) can be reliably detected based on circulating microRNAs. We will then apply these complementary approaches to address fundamental questions in the biology of human diabetes, using both archived material and freshly isolated blood from patients and healthy volunteers. The proposal is close collaboration between three groups with complementary expertise: Schatz (clinical diabetes, biomarkers for early diagnosis), Ferrer (beta cell epigenomics and microRNA) and Dor (beta cell failure and DNA methylation).